J. metamorphic Geol., 2001, 19, 233–248

Formation and exhumation of blueschists and eclogites from NE Oman: new perspectives from Rb–Sr and 40Ar/39Ar dating

A. K. EL-SHAZLY,1 * M. BRO¨ CKER,2 B. HACKER3 AND A. CALVERT3 1Department of Earth Sciences, Sultan Qaboos University, Oman, and Geology Department, Idaho State University, USA ([email protected]) 2Universita¨t Mu¨nster, Institut fu¨r Mineralogie, D-48149 Mu¨nster, Germany 3Geological Sciences Department, University of California, Santa Barbara, CA 93106-9630, USA

ABSTRACT Seven eclogite facies samples from lithologically different units which structurally underlie the Semail ophiolite were dated by the 40Ar/39Ar and Rb–Sr methods. Despite extensive efforts, phengite dated by the 40Ar/39Ar method yielded saddle, hump or irregularly shaped spectra with uninterpretable isochrons. The total gas ages for the phengite ranged from 136 to 85 Ma. Clinopyroxene–phengite, epidote–phengite and whole-rock–phengiteRb–Srisochronsforthesamesamplesyieldedagesof78¡2 Ma.Wethereforeconclude that the eclogite facies rocks cooled through 500 uCatc.78¡2 Ma, and that the 40Ar/39Ar dates can only constrain maximum ages due to the occurrence of excess Ar inhomogeneously distributed in different sites. Our new results lead us to conclude that high-pressure metamorphism of the Oman margin took place in the Late Cretaceous, contemporaneous with ophiolite emplacement. Previously published structural and petrological data lead us to suggest that this metamorphism resulted from intracontinental subduction and crustal thickening along a NE-dipping zone. Choking of this subduction zone followed by ductile thinning of a crustal mass wedged between deeply subducted continental material and overthrust shelf and slope units facilitated the exhumation of the eclogite facies rocks from depths of c. 50 km to 10–15 km within c. 10 Ma, and led to their juxtaposition against overlying lower grade rocks. Final exhumation of all high- pressure rocks was driven primarily by erosion and assisted by normal faulting in the upper plate. Key words: age data; blueschists; eclogites; exhumation; Oman; subduction.

metamorphic rocks from Saih Hatat have so far been INTRODUCTION dated by K–Ar or 40Ar/39Ar methods on phengite Since Michard et al. (1983) and Lippard (1983) reported separates, which yielded ages ranging from 131 to the occurrence of Fe–Mg carpholite-bearing metasedi- 72 Ma (Lippard, 1983; Montigny et al., 1988; El-Shazly ments and blueschists in the Saih Hatat area in NE & Lanphere, 1992; Searle et al., 1994; Miller et al., Oman, at least 10 different tectonic models have been 1999). El-Shazly & Lanphere (1992) interpreted ages as proposed to explain the formation and exhumation of indicating two discrete high-P/T metamorphic events: these high-pressure, low-temperature (high-P/T) meta- an Early Cretaceous event resulting from the sub- morphic rocks. Although these models agree that the duction of the north-eastern part of the Oman continental margin in NE Oman was subducted some- continental margin, and a weaker Late Cretaceous time in the Cretaceous, they disagree on such issues as event related to ophiolite emplacement. Gregory et al. subduction polarity and depth, and the timing of high-P/ (1998) had similar interpretations for the older ages T metamorphism in relation to the emplacement of the (131–82 Ma), but suggested that the continental Semail ophiolite (cf. Lippard et al., 1986; Le Me´tour margin was subducted in a SW-dipping zone before et al., 1990; El-Shazly & Lanphere, 1992; Gregory et al., ophiolite emplacement. On the other hand, Goffe´ et al. 1998; Searle & Cox, 1999). The exhumation mechanism (1988), Le Me´tour et al. (1990), Michard et al. (1994) et al of these high-P/T rocks is also controversial (e.g. and Searle . (1994) argued that phengite from the Oman blueschists and eclogites incorporated excess Ar, El-Shazly & Lanphere, 1992; Michard et al., 1994; and attributed high-P/T metamorphism to Late Chemenda et al., 1996; Miller et al., 1999). Cretaceous subduction of the continental margin of One of the main reasons for the continued con- Oman, either shortly before, or contemporaneous with, troversy over the tectonic history of NE Oman is that the emplacement of the ophiolite. the age of peak metamorphism and the cooling rates of Most of the 40Ar/39Ar white mica spectra reported by the high-P/T rocks are poorly constrained. High-P/T Montigny et al. (1988) and El-Shazly & Lanphere (1992) *Present address: Dept. of Geology, Geography & Physics, Univ. are hump-shaped and do not always satisfy the criteria of Tennessee, TN 38238. Dalrymple & Lanphere (1974) for a meaningful age.

# Blackwell Science Inc., 0263-4929/01/$15.00 233 Journal of Metamorphic Geology, Volume 19, Number 3, 2001 234 A. K. EL-SHAZLY ET AL.

Because other studies of high-pressure rocks have shown of the Hijam Formation, unconformably overlain by Ordovician that incomplete degassing of mica grains, or the quartzites and metasiltstones of the Amdeh Formation (Glennie et al., 40 1974; Le Me´tour et al., 1986). The overlying shelf units consist of incorporation of excess Ar, can produce similar Permian to Lower Cretaceous metacarbonate rocks, interbedded with hump-shaped spectra (e.g. Hacker & Wang, 1995; mafic, arenaceous and pelitic schists. Metamorphosed sandstones, Hannula & McWilliams, 1995; Ruffet et al., 1995), the conglomerates and a me´lange overlie the shelf carbonates either 40Ar/39Ar ages of Montigny et al. (1988) and El-Shazly & unconformably or tectonically, and represent syntectonic foreland basin units, deposited shortly before the final stages of ophiolite Lanphere (1992) may not represent an actual geological emplacement in the Late Cretaceous (Muti Formation; Robertson, event. The internal discordance of the Montigny et al. 1987a, b; El-Shazly, 1995; Fig. 1). 40 39 (1988) Ar/ Ar spectra indicates that their K/Ar ages, El-Shazly & Coleman (1990) subdivided the Saih Hatat area into as well as those of Lippard (1983), cannot be relied upon three thrust-bounded regions based on the structural position of the to indicate the time of the high-pressure metamorphism. lithostratigraphic units exposed in each area and variations in metamorphic grade (Fig. 1). After detailed mapping, Miller et al. Although the white mica 96¡2 Ma age of Searle et al. 40 39 (1999) identified the boundary between Regions II and III as a major (1994) is the only Ar/ Ar age from these eclogites ductile discontinuity separating two plates with different structural which is characterized by a satisfactory plateau, we also and metamorphic features. Their upper plate therefore includes consider this age suspect, because every other sample of imbricated, unmetamorphosed to weakly metamorphosed units, and these high-pressure rocks has yielded uninterpretable corresponds to Regions I and II of El-Shazly & Coleman (1990), 40 39 whereas their lower plate consists of strongly folded and deeply Ar/ Ar spectra (including the other two published by metamorphosed units, and corresponds to Region III (Fig. 1). Searle et al., 1994). The units of Region I, located between Dar Sait and Hamiriya (the Recent studies have clearly shown that the 40Ar/39Ar southern suburb of Ruwi) in northern Saih Hatat (Fig. 1), include data of white mica from high-pressure rocks are often unmetamorphosed Hawasina cherts, serpentinites and basalts, thrust onto a tectonic me´lange (the Ruwi me´lange). Following the contaminated with excess argon (e.g. Arnaud & Kelley, interpretations of Le Me´tour et al. (1986) and Robertson (1987a, 1995; Ruffet et al., 1995; Reddy et al., 1996). U–Pb b), the Ruwi me´lange is herein considered as the metamorphic dating of zircon inclusions in garnet and Sm–Nd dating equivalent of some members of the Muti Formation. of garnet from the Dora Maira eclogite facies rocks Region II consists of several thrust sheets of folded (parautochtho- (western Alps) both yielded Tertiary ages (e.g. Tilton nous) basement and shelf units (Precambrian to Early Cretaceous), metamorphosed under lawsonite–albite to epidote–blueschist facies et al., 1991; Gebauer et al., 1997) that cast doubt on the conditions. These units exhibit a downsection increase in meta- 40 39 previously accepted Cretaceous Ar/ Ar plateau ages morphic grade (El-Shazly, 1994, 1995, 1996). In the northern parts of obtained for phengite from the same rocks (e.g. Region II, the Early Cretaceous units are missing, as the Ruwi Hunziker et al., 1989). Schermer et al. (1990), Tonarini me´lange (Region I) is tectonically juxtaposed against Jurassic and Triassic units, all of which are deformed into a series of tight, et al. (1993) and Li et al. (1994) reported Rb–Sr ages on overturned folds verging to the south or south-west. In the southern phengite from Mount Olympos blueschists, Greece, and western parts of Region II, the Hatat schists, Hijam dolomites, Himalayan eclogites, Pakistan, and Dabie Shan Amdeh quartzites and Saiq carbonates are deformed into large-scale eclogites, China, that are younger than their corre- overturned to recumbent open folds, and smaller scale cylindrical to sponding 40Ar/39Ar ages, although the phengite closure sheath-like folds (Le Me´tour et al., 1986; Gregory et al., 1998; Miller 40 39 et al., 1999). A NE–SW-stretching lineation is well developed within temperature for Ar/ Ar is lower than that for Rb–Sr. the Hatat schists (Le Me´tour et al., 1986; Michard et al., 1994; These results underscore the need for the application of Gregory et al., 1998; Miller et al., 1999). Miller et al. (1999) also other geochronological techniques alongside 40Ar/39Ar report rare shear bands that record a top-to-the-NE transport dating for determining the ages of eclogites and direction in the deepest units of Region II. Region III, the structurally lowest region, is more coherent, blueschists, and making meaningful tectonic interpre- consisting of calcareous mica schists, metacarbonate rocks, mafic tations (e.g. Sherlock et al., 1999). schists, metapelites and quartz–mica schists (Fig. 1). All of these units In this paper, we present new whole-rock and mineral were considered by Le Me´tour et al. (1986) as part of the Permian separate Rb–Sr data from the As-Sifah eclogite facies Saiq Formation, although 13C data do not support a Permian age for rocks in eastern Saih Hatat. We also present new the protolith of some calcareous schists (Gregory et al., 1998). An 40 39 eastward increase in metamorphic grade and intensity of deformation Ar/ Ar data on phengite from the same samples, this led El-Shazly et al. (1990) to identify three metamorphic zones based time using replicate isothermal step heating to help primarily on mineral assemblages in mafic schists. In zone A, the remove or identify any possible excess Ar. This study is mafic unit is characterized by greenschists, crossite–epidote schists therefore aimed at: (i) constraining the age(s) of high-P/T and minor clinopyroxene-rich granofelses, interlayered on a centi- metre to decimetre scale. In zone B, garnet–glaucophane schists are metamorphism; (ii) providing insight into the cooling interlayered with glaucophane–epidote schists, clinopyroxene-rich rates of these rocks; and (iii) presenting a revised tectonic granofelses and epidote amphibolites. Zone C is defined by the model for the evolution of the Saih Hatat area in NE appearance of eclogites that are interlayered with garnet blueschists, Oman. clinopyroxene-rich granofelses and epidote–amphibolites (El-Shazly et al., 1997). Peak conditions for this zone are estimated at 550–580 uC and 12–15 kbar (El-Shazly et al., 1997; El-Shazly, GEOLOGICAL SETTING 2001), although Wendt et al. (1993) and Searle & Cox (1999) favoured much higher pressures of 20 kbar for these rocks. A detailed The Saih Hatat area in NE Oman is a domal window that exposes description of the mineral assemblages of the different metamorphic basement, shelf and foreland basin units structurally beneath the zones of Region III was presented in El-Shazly et al. (1990), El-Shazly allochthonous Semail ophiolite, Haybi Complex and Hawasina basin & Liou (1991), El-Shazly et al. (1997) and El-Shazly (2001). units (Glennie et al., 1974; Fig. 1). The Saih Hatat basement consists The units of Region III experienced at least three deformational of Proterozoic to Precambrian quartz–mica schists, metagreywackes events (e.g. Searle et al., 1994; Miller et al., 1999). The earliest and metabasites of the Hatat Formation and recrystallized dolomites deformation resulted in a penetrative foliation that becomes layer FORMATION and EXHUMATION OF OMAN ECLOGITES 235

Fig. 1. Geological map of Saih Hatat, NE Oman, simplified after Le Me´tour et al. (1986) and Gregory et al. (1998). Regions I, II and III and metamorphic zones A, B and C are from El-Shazly & Coleman (1990) and El-Shazly et al. (1990). The key on the right represents the allochthonous units, whereas the key on the left represents the autochthonous and parautochthonous basement, shelf and foreland basin units. Q, Quaternary; T, Tertiary; K, Cretaceous; Tr, Triassic; J, Jurassic; P, Permian; ps, Precambrian. Inset: sample location map showing the mafic unit (hatched), quartz–mica schists (stippled) and calcareous schists (unpatterned). Note that most of the major structural boundaries shown here as thrusts have been reactivated as extensional shear zones during exhumation. parallel in zones B and C, and is axial planar to later folds. The second have so far been the most controversial. Moreover, constraining the deformational event (D2) produced the N to NE-vergent folds (which ages of the highest grade metamorphic rocks in Saih Hatat is essential are inclined in zone A and become recumbent sheath folds in zone C), for tectonic interpretations. Seven samples collected from two coastal and which folded the D1 foliation. A NNE-trending lineation which outcrops c. 1 km north of As-Sifah (Fig. 1) were selected for Rb–Sr plunges to the NNE or SSW (Michard et al., 1984; Le Me´tour et al., and 40Ar/39Ar age dating after careful petrographic analysis with a 1990; Gregory et al., 1998) was also produced by D2. The third polarizing microscope and, in some cases, microprobe analysis of the deformation (D3) took place under greenschist facies conditions and constituent minerals. These samples were collected from three distinct resulted in S–C and rotational fabrics and extensional crenulation rock types interlayered on a decimetre scale: mafic blueschists, Cpx- cleavages (El-Shazly & Lanphere, 1992; Searle et al., 1994; Miller et al., rich granofelses and metasedimentary schists (Table 1). All layers are 1999). A possible fourth deformational event is observed locally cofacial and have had identical P–T histories (cf. El-Shazly et al., in zones B and C, and is manifested by broad, asymmetrical folds 1997). An eighth sample (C-106) of a boudinaged monomineralic vein with E–W-striking axial planes, and a top-to-the-N? sense of of coarse-grained phengite cutting a retrogressed eclogite boudin was shear. The development of late NW-striking, SW-dipping veins of collected for 40Ar/39Ar dating from an outcrop c. 500 m west of the Qz¡Cc¡Hm¡Ab¡Ep¡Chl, typicallywithgreenschistfaciesaltera- coastal outcrops. Unlike the rest of the samples, this vein is likely to tion halos cutting across lower plate units, marks a later brittle have formed during the early stage of exhumation of the blueschists deformation. Localized, east-directed rockslides and falls represent the and eclogites of zone C. latest deformation and appear to be active to the present day. The seven eclogite facies samples were crushed and sieved, and phengite was separated from the 500–1000 mm size fraction using an isodynamic separator at Sultan Qaboos University, shaking on filter SAMPLING, PETROGRAPHY AND MINERAL paper, and hand-picking under a binocular microscope. An CHEMISTRY additional aliquot of phengite was separated from the 250–500 mm size fraction for samples C-64, C-52 and C-62 to investigate the effect of grain size on the 40Ar/39Ar age spectra. Crystals with inclusions Sample selection and analytical techniques were eliminated, and all separates were optically homogeneous. Each In this study, we focus on dating samples from the highest grade zone separate was divided into three aliquots: two for Rb–Sr and 40Ar/39Ar C of Region III. Ages obtained on eclogite facies rocks from this zone dating, and the third for scanning electron microscopy (SEM) 236 A. K. EL-SHAZLY ET AL. imaging and analysis. The aliquot used for Rb–Sr dating was further On the other hand, C-62 was significantly affected by retrogression cleaned by grinding in ethanol using an agate mortar, and washing in where garnet is partially replaced by phengite, epidote, glaucophane, ethanol (P.a.) and H2O (three times distilled) in an ultrasonic bath. albite and chlorite along fractures and rims, and symplectites of The aliquot used for SEM analysis was mounted on a glass slide with albite+sodic–calcic amphibole (replacing Cpx?) are common. epoxy, polished, carbon coated and analysed using a JEOL 840A Phengites are medium-grained (1.5–2 mm), subidioblastic crystals scanning electron microscope at Sultan Qaboos University. Details of that define the foliation and are devoid of kink bands. In C-66 and C- standardization and analysis are described in El-Shazly (2001). 79, phengite contains a few inclusions of rutile (¡minor garnet), very few fractures and is regularly zoned from Si- and Mg-rich cores to less siliceous rims with higher Fe/Mg (Fig. 2c; Table 2). In C-62, the phengite is complexly and irregularly zoned, has a higher density of Petrography and mineral chemistry fractures and is occasionally rimmed by chlorite (Table 2). There are no chemical differences between the different size fractions of Mineral assemblages are listed in Table 1. Samples C-52 and C-67 are phengite separated from this sample. mafic blueschists consisting of predominantly Gln+Ep+Ph, with Phengite of sample C-106 is very coarse grained (1–2 cm long). It is smaller amounts of Cpx¡Gt. In both samples, glaucophane and almost chemically homogeneous with Si=3.53–3.56 apfu and the epidote are in textural equilibrium with phengite, but appear to have lowest Fe/Mg ratios (0.17–0.45) in all dated samples (Table 2). It has crystallized after (and possibly at the expense of) Gt¡Cpx. Retro- ¡ ¡ very few fractures, but contains many inclusions of albite, biotite, grade phases include Ab+Cc+Chl Win Bt, and are more chlorite, rutile and winchite near its rims. common in C-52 (Table 1). Phengite occurs as fine- to medium- grained crystals (0.4–2 mm long) that define the foliation. The crystals are not kinked, and appear to have been partially polygonized during D2. Phengite contains numerous inclusions of 40Ar/39Ar DATING rutile, and rare inclusions of Hm¡Ttn¡Ep. In C-52, the phengite crystals are significantly fractured, with some fractures filled with Analytical techniques albite, chlorite and winchite. Phengite from C-52 is irregularly but weakly zoned (Fig. 2a), with both separated size fractions having the Phengite separates were wrapped in Cu foil, sealed in a quartz vial same composition and zoning. In C-67, phengite crystals are regularly and irradiated at the TRIGA reactor at Oregon State University for zoned from Si- and Mg-rich cores (Si=3.40–3.62 atoms per formula 16 MWh. Ratios of reactor-produced Ca-derived isotopes were unit (apfu), calculated on the basis of 11 oxygens; Fe/Mg=0.47–0.56) established by analysing CaF2 included in the irradiation vial. to less siliceous rims (Si=3.30 apfu; Fe/Mg=0.62–0.88; Table 2). Sanidine from the Fish Canyon tuff, with a preferred age of 27.8 Ma, Samples C-60 and C-64 were collected from eclogitic Cpx-rich was used as a neutron flux monitor. The monitor packets were layers. Both samples are characterized by a weak foliation defined by interspersed between every six unknown packets; c. 0.8 mg of each the orientation of Cpx¡Ph, and are pristine, containing only traces monitor was analysed with the resistance furnace to determine of retrograde albite, winchite, chlorite¡biotite (Table 1). Phengite irradiation parameter, J. We used an uncertainty of 0.2% (at 1s) for J occurs as fine- to medium-grained (0.6–2 mm) subidioblastic crystals for all samples. which appear to have crystallized after garnet and Cpx, often in All samples, which are multigrain splits of small mass (c. 1 mg), garnet pressure shadows. In C-60, phengite is almost devoid of were analysed over a two-week period using a double-vacuum inclusions, and is regularly zoned from Si- and Mg-rich cores Staudacher-type resistance furnace. To improve upon previous (Si=3.60 apfu, Fe/Mg=0.4) to less siliceous, more ferroan rims efforts at dating phengite, and in an attempt to resolve the influence (Si=3.38 apfu, Fe/Mg=0.5–0.9; Table 2). In C-64, phengite crystals of excess Ar and complications in spectral interpretation, we are more fractured and contain several inclusions of Cpx and rutile, performed a large number of isothermal heating steps. Each cycle and are sometimes partially rimmed by albite. The zoning patterns comprised 2 min of heating from 300 uC to the set temperature, are more irregular, although they show an overall trend of decreasing 5–90 min (most commonly 10 min) of constant temperature, and Si and increasing Fe/Mg from core to rim (Fig. 2b; Table 2). 3 min cooling to 300 uC. Gas was gettered continuously during Different size fractions of phengite from this sample have similar extraction with two SAES ST172 Zr–V–Fe getters operated at 4 A compositions and zoning patterns. and 2 A. The collected gas was analysed for 7–9 min in an MAP-216 Samples C-62, C-66 and C-79 were all collected from metasedi- spectrometer with a Baur–Signer source and Johnston MM1 electron mentary schists which contain i 30 vol.% phengite, as well as multiplier operating in static mode with a room temperature SAES significant amounts of garnet and glaucophane (Table 1). These rocks ST707 Zr–V–Fe getter. Peak heights at the time of gas introduction differ from the Na- and Fe-rich metapelites described by El-Shazly & into the mass spectrometer were determined by extrapolating the Liou (1991), as they lack paragonite and chloritoid (Table 1). Their evolved signal size using a linear regression. The mass discrimination, most likely protolith is a mixture of mafic to intermediate tuffs and determined by analysing the 40Ar/36Ar ratio of air, was 0.993 per mudstones. Sample C-79 is pristine, and almost devoid of any signs of atomic mass unit (a.m.u.) during the course of these experiments. retrogression. Sample C-66 shows a few retrograde features such as Resistance furnace m/e 40 blanks varied from 2.9310x16 mol at glaucophane and Cpx partially replaced by oxychlorite and calcite. 800 uC to 1.2310x15 mol at 1400 uC.

Table 1. Modal contents of samples dateda.

Sample Gt Cpx Gln Ep Ph Chl Ab Qz Opq Rt Others

Mafic blueschists C-52 5 3 15–20 30 10–15 5–7 15 2 2 tr Cc(10), Win(2), Ttn(tr) C-67 — 20 20–25 20–25 15 3–5 3–5 — 7–10 2–3 Cc(3–5), Bt(tr)

Cpx-rich C-60 10–12 50–60 5 3 15 tr tr 3–5 2–3 1 Ap(tr) C-64 20–25 45–50 3 Aln(3) 7–10 1–2 tr 7–10 Mgt(1–2) tr Bt(tr)

Metasedimentary schists C-62 10–15 tr 20 2 30 5–7 7–10 2–3 7–10 tr Cc(2–3), Aln(tr), Win(2) C-66 1–2 7–10 10–15 2–3 55–60 5 3 5 3 — Cc(5), Ap(tr) C-79 15–20 2–3 20–25 2–3 30 3–5 — 5 1–2 2 Cc(15–20), Bt(tr) aModal contents estimated visually. tr, traces. FORMATION and EXHUMATION OF OMAN ECLOGITES 237

Fig. 2. Sketches of representative phengite grains from samples (a) C-52 (grain 1), (b) C-64 (grain 8) and (c) C-79 (grain 4), showing their Si contents in atoms per formula unit calculated on the basis of 11 oxygen atoms. Densely stippled grains, inclusions of rutile; light stipple, titanite; diagonal rules, chlorite; vertical rules, epidote.

steps. Apparent ages given by the intermediate Results temperature steps increase gradually to some peak The spectra obtained for all samples are characterized value before dropping once more, giving rise to a ‘W’- by K/Ca (>0.04) and K/Cl ratios that are within the or saddle-shaped spectrum with an intermediate hump normal range for K-white mica, suggesting that the (Figs 3 & 4). The spectrum for C-67 is overall saddle- separates are indeed pure. Nevertheless, these samples shaped with some irregularities (Fig. 5); 45% of the gas yield complex spectra with different shapes and total defines a plateau age of c. 100 Ma. Samples C-52 gas ages that range from 148 to 84.9 Ma (Table 3). The (Fig. 5), C-60 (Fig. 3) and C-79 (Fig. 4) yield hump- simplest spectrum, aliquot C-62f, decreases from low-T shaped spectra with several irregularities. steps at c. 95 Ma to high-T steps at c. 92 Ma. The In the three samples where two aliquots of different coarser grained aliquot of this sample, C-62c, yields a grain sizes were dated, the finer grained size fractions weakly hump-shaped spectrum from 96 to 92 Ma. always yielded slightly younger apparent total gas ages Samples C-64, C-66 and C-106 all yield the oldest (Table 3). Extensive isochron analysis does not reveal apparent ages in their lowest and highest temperature systematic behaviour for any of the spectra, and we

Table 2. Representative analyses of phengite.

C-52 C-60 C-62 C-64 C-66 C-79 C-67 C-106 Sample 1 6 1 4 7 1 1 2 2 2c 4 2 Grain 8 9 11 21 1 3 1 8 5 14 1 2 2 3 4 1 4 5 11 20 7 1 7 5 16 19 22 Analysis c r c r c r r c c r c r c r c r c r r c r r c c r r ir ciror

SiO2 53.25 53.61 54.23 53.97 52.72 51.50 51.32 50.57 53.49 50.69 51.88 50.91 51.29 51.85 52.33 52.34 52.81 52.08 49.78 50.80 52.92 51.59 51.95 51.96 48.92 49.53 50.78 54.05 53.16 50.35 TiO2 0.27 0.25 0.17 0.18 0.16 0.20 0.23 0.23 0.21 0.25 0.20 0.33 0.26 0.31 0.21 0.29 0.25 0.29 0.28 0.15 0.26 0.43 0.25 0.17 0.37 0.35 0.30 0.19 0.32 0.22

Al2O3 23.51 24.25 22.47 24.14 20.61 22.34 23.09 22.36 22.56 24.62 23.30 23.73 23.70 23.45 19.73 23.46 21.07 22.20 25.89 23.91 21.08 24.86 19.93 20.53 26.50 26.75 23.33 22.46 21.43 21.52 Cr2O3 0.00 0.00 0.01 0.00 0.00 0.00 0.10 0.00 0.01 0.00 0.06 0.05 0.00 0.00 0.03 0.03 0.04 0.00 0.03 0.06 0.06 0.01 0.09 0.15 0.13 0.14 0.00 0.11 0.08 0.13 FeO 2.66 2.78 2.29 2.80 3.76 3.67 4.66 4.56 4.76 4.46 3.74 3.64 3.91 3.90 5.87 4.32 5.52 4.76 4.66 4.59 4.85 4.40 4.84 4.44 4.34 3.95 4.52 2.42 2.54 5.68 MnO 0.00 0.04 0.00 0.00 0.05 0.03 0.00 0.03 0.00 0.00 0.00 0.47 0.00 0.04 0.01 0.05 0.01 0.01 0.00 0.03 0.00 0.03 0.00 0.00 0.02 0.06 0.06 0.00 0.00 0.00 MgO 4.62 4.43 5.00 4.47 4.99 4.69 3.76 3.76 4.08 3.23 4.25 4.02 4.09 4.19 4.41 4.12 4.26 4.17 2.85 4.24 4.44 4.23 4.83 4.99 2.90 2.78 4.10 5.31 5.27 6.96

Na2O 0.44 0.42 0.35 0.45 0.28 0.40 0.33 0.26 0.30 0.46 0.38 0.42 0.42 0.36 0.27 0.34 0.19 0.37 0.68 0.42 0.28 0.37 0.20 0.35 0.91 0.92 0.47 0.31 0.26 0.20 K2O 10.88 10.76 10.75 10.82 10.84 10.92 10.93 10.79 11.29 10.58 11.16 11.35 11.29 11.38 11.06 11.00 11.51 11.05 10.43 10.86 11.29 10.89 11.30 11.04 10.59 10.53 11.16 11.41 11.16 9.56 Total 95.63 96.54 95.28 96.83 93.42 93.75 94.42 92.57 96.70 94.30 94.98 94.91 94.96 95.48 93.93 95.95 95.65 94.93 94.60 95.06 95.19 96.81 93.39 93.63 94.68 95.01 94.72 96.25 94.22 94.62

Si 3.55 3.54 3.61 3.55 3.63 3.54 3.51 3.53 3.57 3.47 3.53 3.47 3.49 3.51 3.63 3.51 3.59 3.55 3.40 3.46 3.60 3.44 3.61 3.59 3.34 3.36 3.47 3.59 3.61 3.45 AlIV 0.43 0.45 0.38 0.44 0.37 0.45 0.48 0.46 0.42 0.52 0.46 0.51 0.49 0.48 0.36 0.47 0.40 0.44 0.59 0.54 0.39 0.54 0.38 0.40 0.64 0.62 0.51 0.40 0.37 0.54 Ti 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.02 0.02 0.02 0.01 0.02 0.01 Al(total) 1.85 1.89 1.77 1.87 1.67 1.81 1.86 1.84 1.78 1.98 1.87 1.91 1.90 1.87 1.61 1.86 1.69 1.78 2.08 1.92 1.69 1.95 1.63 1.67 2.13 2.14 1.88 1.76 1.72 1.74

AlVI 1.41 1.44 1.39 1.43 1.31 1.36 1.39 1.38 1.36 1.46 1.40 1.39 1.41 1.39 1.25 1.39 1.29 1.35 1.50 1.38 1.30 1.41 1.25 1.27 1.49 1.52 1.37 1.36 1.35 1.20 Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.01 0.00 0.01 0.00 0.01 Fe 0.15 0.15 0.13 0.15 0.22 0.21 0.27 0.27 0.27 0.26 0.21 0.21 0.22 0.22 0.34 0.24 0.31 0.27 0.27 0.26 0.28 0.25 0.28 0.26 0.25 0.22 0.26 0.13 0.14 0.33 Mn 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Mg 0.46 0.44 0.50 0.44 0.51 0.48 0.38 0.39 0.41 0.33 0.43 0.41 0.41 0.42 0.46 0.41 0.43 0.42 0.29 0.43 0.45 0.42 0.50 0.51 0.30 0.28 0.42 0.53 0.53 0.71 Na 0.06 0.05 0.05 0.06 0.04 0.05 0.04 0.04 0.04 0.06 0.05 0.06 0.06 0.05 0.04 0.04 0.03 0.05 0.09 0.06 0.04 0.05 0.03 0.05 0.12 0.12 0.06 0.04 0.03 0.03 K 0.93 0.91 0.91 0.91 0.95 0.96 0.95 0.96 0.96 0.92 0.97 0.99 0.98 0.98 0.98 0.94 1.00 0.96 0.91 0.94 0.98 0.93 1.00 0.97 0.92 0.91 0.97 0.97 0.97 0.84 Oct total 2.02 2.03 2.01 2.02 2.04 2.05 2.04 2.04 2.03 2.05 2.05 2.04 2.04 2.04 2.05 2.05 2.04 2.04 2.05 2.08 2.03 2.08 2.03 2.05 2.05 2.03 2.05 2.02 2.03 2.25 Fe/Mg 0.32 0.35 0.26 0.35 0.42 0.44 0.70 0.68 0.65 0.78 0.49 0.51 0.54 0.52 0.75 0.59 0.73 0.64 0.92 0.61 0.61 0.60 0.56 0.51 0.83 0.79 0.62 0.26 0.27 0.46 c, core; r, rim; ir, inner rim; or, outer rim. 238 A. K. EL-SHAZLY ET AL.

Fig. 4. 40Ar/39Ar age spectra for phengite from Fig. 3. 40Ar/39Ar age spectra for phengite from Cpx-rich metasedimentary samples C-62, C-66 and C-79. C-62f is the samples C-60, C-64 and vein C-106. C-64f is the finer grained finer grained aliquot of sample C-62. aliquot of sample C-64. corrected for mass fractionation using a factor deduced from multiple interpret the lowest ages in each spectrum to be a measurements of Rb standard NBS 607. Total procedural blanks maximum age for the sample. were < 0.1 ng for Rb and < 0.15 ng for Sr. Based on repeated measurements, the 87Rb/86Sr ratios were assigned an uncertainty of 1% (2s). For other isotope ratios, uncertainties are reported at the 2sm level. Repeated runs of NBS standard 987 gave an average Rb–Sr DATING 87Sr/86Sr ratio of 0.710265¡24 (2s, n=41). All ages and elemental concentrations were calculated using the IUGS recommended decay constants (Steiger & Ja¨ger, 1977). Rb–Sr ages were calculated using Analytical techniques the least-squares regression technique of York (1969). Ages and Isotope analyses were carried out at the Zentrallaboratorium fu¨r errors are reported at the 2s level. The Rb–Sr isotopic data are Geochronologie at the Institut fu¨r Mineralogie, Universita¨t Mu¨nster, summarized in Table 4. using a VG Sector 54 multicollector mass spectrometer (Sr) and an NBS-type Teledyne mass spectrometer (Rb). For Rb–Sr analyses, whole-rock powders (about 100 mg) and mineral separates (phengite, c. 24–43 mg; epidote, c. 3–5 mg; omphacite, c. 95 mg) were mixed Results with an 87Rb–84Sr spike in Teflon screw-top vials and dissolved in an Two-point isochrons using phengite–whole rock, HF–HNO3 (5 : 1) mixture on a hot plate overnight. After drying, 6 N HCl was added to the residue. This mixture was homogenized on a phengite–epidote or phengite–clinopyroxene yield hot plate overnight. After a second evaporation to dryness, Rb and Sr ages between 77.8¡1.2 and 77.2¡0.8 Ma for all were separated by standard ion-exchange procedures (AG 50W-X8 mafic blueschist and metasedimentary schist samples resin) on quartz glass columns using 2.5 N and 6 N HCl as eluents. (Table 4). Isochrons for the Cpx-rich samples (C-60 For mass spectrometric analysis, Rb and Sr were loaded on Ta ¡ filaments with H2O and H3PO4, respectively. Correction for mass & C-64) yield slightly older ages of 79.2 0.8 fractionation is based on an 86Sr/88Sr ratio of 0.1194. Rb ratios were and 78.6¡0.9 Ma, respectively (Fig. 6; Table 4).

Table 3. Summary of results of 40Ar/39Ar dating of phengite.

Sample Size fraction (mm) J Total gas age (Ma) Preferred age (Ma) Shape of spectrum

C-52f 250–500 0.0039992 125 j 108 Hump C-52 500–1000 0.0039699 131 j 115 Hump C-67 500–1000 0.0040127 99.4 j 96 Saddle+hump

C-60 500–1000 0.0040022 111 j 108 Hump C-64f 250–500 0.0039898 131 j 130 Hump C-64 500–1000 0.0039962 136 j 133 Saddle+hump

C-62f 250–500 0.0039724 93.6 j 92 Irregular C-62 500–1000 0.0040090 94.1 j 92 Irregular C-66 500–1000 0.0039930 98.4 j 96 Saddle+hump C-79 500–1000 0.0039785 84.9 j 83 Hump C-106 >1 cm 0.0039865 148 j 146 Saddle FORMATION and EXHUMATION OF OMAN ECLOGITES 239

Table 4. Results of Rb–Sr dating.

Sample Type Rb (ppm) Sr (ppm) 87Rb/86Sra87Sr/86Sr Errorb Age (Ma)

C-52 Whole rock 17.49 126.3 0.401 0.713691 22 C-52 Epidotec 4.585 466.7 0.028 0.713386 24 77.2¡0.8 C-52 Phengite 274.0 6.769 118.651 0.843512 33 C-67 Whole rock 48.43 52.99 2.645 0.712511 23 C-67 Epidote 2.575 870.0 0.0086 0.711909 22 77.6¡0.8 C-67 Phengite 432.9 9.575 132.750 0.858253 27 C-60 Whole rock 94.08 25.36 10.747 0.720569 21 78.6¡0.9 C-60 Phengite 397.9 14.13 82.21 0.800357 34 C-64 Whole rock 21.52 58.50 1.064 0.711774 24 C-64 Omphacite 2.835 18.42 0.445 0.710679 22 79.2¡0.8 C-64 Phengite 287.9 10.38 80.982 0.801284 27 C-62 Whole rock 137.3 39.26 10.128 0.720599 21 C-62 Epidote 1.771 565.2 0.0091 0.710344 21 77.6¡0.8 C-62 Phengite 593.1 8.181 214.640 0.947072 41 C-66 Whole rock 190.3 43.81 12.587 0.723770 24 77.8¡1.2 C-66 Phengite 430.5 8.793 143.899 0.868834 39 C-79 Whole rock 150.3 49.31 8.830 0.722599 20 C-79 Epidote 3.413 614.4 0.016 0.713796 20 77.4¡0.8 C-79 Phengite 336.1 3.305 303.985 1.047872 44

aThe 87Rb/86Sr ratios were assigned uncertainties of 1% 2s. b 87 86 Values indicate uncertainties in the last two digits of the Sr/ Sr ratios quoted at the 2sm level. cItalics indicate phases used for age calculation. Fig. 5. 40Ar/39Ar age spectra for phengite from mafic samples C-67 and C-52. C-52f is the finer grained aliquot of sample resetting of the Rb–Sr system and the perturbation of C-52. the 40Ar/39Ar spectra. 2 The younger Rb–Sr ages result from the mobilization 87 86 ( Sr/ Sr)i ratios range from 0.709 to 0.714, and show of Sr by fluids during retrogression, and its subsequent no correlation with rock type or extent of retrogression. loss from the dated minerals, whereas Ar, being an inert Whole-rock Rb values for the dated samples range gas, was unaffected by these retrogressive fluids. from 17.49 to 190 ppm, whereas the 87Rb/86Sr ratios 3 The Rb–Sr ages of 78¡2 Ma represent the time at range from 0.4 to 12.58 (Table 4). which the As-Sifah eclogite facies rocks cooled through 500¡50 uC, whereas the 40Ar/39Ar ages of i 83 Ma can only be considered maximum ages of peak INTERPRETATION OF THE AGE DATA metamorphism, as this phengite has incorporated The Rb–Sr ages defined by two-point isochrons are excess Ar. It is unlikely that the observed difference in similar for all samples, regardless of the lithology or the 40 39 extent of retrogression of the dated samples. On the Ar/ Ar and Rb–Sr ages is representative of two other hand, the 40Ar/39Ar age spectra are all perturbed, metamorphic events, because the closure temperature 40 with different samples yielding different total gas of phengite for Sr is higher than that for Ar (e.g. ages. As in the case of the Rb–Sr ages, there is no Purdy & Ja¨ger, 1976; Blanckenburg et al., 1989). Loss relationship between the apparent 40Ar/39Ar ages of Sr from phengite during retrogression is also an yielded by each phengite separate and the rock type unlikely explanation, simply because different samples or its extent of retrogression. There is also no clear with different degrees of retrogression (e.g. C-79 & C- relationship between the chemistry of the phengite (e.g. 62) yield the same Rb–Sr ages. Therefore, the most the Si contents, zoning patterns or Fe/Mg ratios) and plausible explanation for the age difference is that the the apparent ages (cf. Scaillet et al., 1992). Rb–Sr ages record cooling shortly after a period of Given that the ideal closure temperature for Rb–Sr extensive crystallization close to 78¡2 Ma, whereas 40 39 in phengite (c. 500¡50 uC; Purdy & Ja¨ger, 1976; the Ar/ Ar ages are not meaningful because of the Blanckenburg et al., 1989) is higher than that for Ar in incorporation of excess Ar in phengite. the same mineral (350–430 uC; e.g. Purdy & Ja¨ger, This conclusion is supported by several observations. 1976; Sisson & Onstott, 1986; Blanckenburg et al., 1 There is large variability in 40Ar/39Ar spectral shapes 1989; Kirschner et al., 1996), the 40Ar/39Ar ages are and total gas ages from samples with identical P–T expected to be younger than the Rb–Sr ages for the histories. same sample. However, the Rb–Sr ages (78.2¡2 Ma) 2 The oldest apparent 40Ar/39Ar age is recorded by C- are considerably younger than any of our 40Ar/39Ar 106 phengite. However, this sample is younger than all ages (146–83 Ma). This could have three possible other eclogitic phengite, as it formed during retro- explanations. gression and veining. 1 The 40Ar/39Ar ages approximate the age of the high- 3 Different grain size separates from the same sample P/T peak metamorphism, whereas the 78¡2 Ma age yield different 40Ar/39Ar ages. The fact that the coarser represents a second metamorphic event that caused the aliquots yield older ages leads us to conclude that 240 A. K. EL-SHAZLY ET AL.

Fig. 6. 87Sr/86Sr–87Rb/86Sr isochrons for samples C-62, C-66, C-52, C-67, C-79, C-64 and C-60. Note that most error bars are smaller than the symbol sizes (Table 4). excess Ar was added before the most recent Ar loss (except C-62), our separates consist of a single event. If the phengite underwent a growth event or generation of phengite, and there is no relation between complete recrystallization such that all Ar was lost and the extent of sample retrogression or zoning patterns of then the excess Ar was added diffusively to the crystals, these micas and their resulting age spectra. the finer grained samples would have been most 5 Phengite with a higher density of fractures (e.g. C-64 affected and would have yielded the oldest apparent & C-52) yields older apparent ages compared to that ages, which is not the case. from less fractured crystals. This is consistent with the 4 Saddle-shaped 40Ar/39Ar spectra, similar to those of incorporation of excess Ar, a significant component of C-67, C-64 and C-106, are considered to be indicative some metamorphic fluids (e.g. Cumbest et al., 1994), of the presence of excess Ar (Lanphere & Dalrymple, via fractures in phengite (e.g. Reddy et al., 1996). 1976; McDougall & Harrison, 1988). Although more The results of this study are not unique. Dallmeyer difficult to interpret, hump-shaped spectra, similar to et al. (1990) reported variably discordant 40Ar/39Ar age those of C-52, C-60 and C-79, have also been attributed spectra for phengite from blueschists and eclogites to excess Ar (e.g. Hannula & McWilliams, 1995). Other from west-central Spitsbergen that yielded slightly interpretations of hump-shaped spectra as dating a older ages compared to those obtained from Rb–Sr mixture of two populations of white mica with different phengite–whole-rock isochrons. Tonarini et al. (1993) ages (Wijbrans & McDougall, 1986), partial re- reported 40Ar/39Ar plateau ages for phengite that were equilibration of celadonite-rich phengite under lower considerably older than Sm–Nd and Rb–Sr isochron pressure during exhumation (Scaillet et al., 1992) and ages for eclogites from the Himalayas. Li et al. (1994) subgrain chemical heterogeneity (Hammerschmidt & and Inger et al. (1996) also reported much older Franz, 1992) are less likely because, for all samples 40Ar/39Ar ages than corresponding Rb–Sr phengite FORMATION and EXHUMATION OF OMAN ECLOGITES 241 ages from Dabie Shan, China and the Sesia Lanzo (1994) and Searle et al. (1994) that the As-Sifah units eclogites, western Alps, respectively. Bro¨cker & Franz represent a terrane different from the remaining units (1998) and Sherlock et al. (1999) reported younger Rb– in the lower plate, or that they represent Hawasina Sr ages for blueschist facies phengite compared to their continental slope units metamorphosed under high-P/T corresponding 40Ar/39Ar ages for samples from the conditions (Oberha¨nsli et al., 1999), are not supported intermediate unit of Tinos, Cyclades, and the Tavsanli by conclusive data. zone, north-west Turkey, respectively. Along with the 2 The most likely protolith of the Ruwi tectonic results of Tilton et al. (1991) and Gebauer et al. (1997) me´lange (Region I) is one of the members of the Late on Dora Maira eclogites, it seems that excess Ar is Cretaceous Muti Formation (Le Me´tour et al., 1986; common in high-pressure phengite. The amount of Robertson, 1987a, b; El-Shazly, 1994). The suggestion excess Ar apparently increases with increased access to of Searle et al. (1994) that this me´lange is part of the fluids and increasing metamorphic pressures (e.g. Haybi Complex is less likely. The Haybi Complex (a Sherlock et al., 1999). mixture of volcanics, olistostromes and exotic lime- stones) is always thrust on top of the unmetamor- phosed Hawasina sediments (Searle et al., 1980; Searle TECTONIC SYNTHESIS & Cooper, 1986), whereas the Ruwi me´lange is The success of any tectonic model for the formation sandwiched between the shelf units and the overthrust and exhumation of the high-P/T metamorphic rocks of Hawasina cherts and serpentinites (Le Me´tour et al., Saih Hatat is gauged by its ability to account for the 1986; El-Shazly, 1994). stratigraphic, geochronological, petrological, cooling 3 The bulk of the igneous crustal rocks of the Semail rate and structural data available for this area. More ophiolite crystallized between 97 and 94 Ma, as than 10 different models have been proposed for the indicated by U–Pb zircon ages on plagiogranites tectonic evolution of Saih Hatat. The principal (Tilton et al., 1981), 40Ar/39Ar hornblende ages on differences among these models pertain to the timing plagiogranites, gabbros and veins (Hacker et al., 1997), of high-P/T metamorphism and the peak pressures and radiolarian and foraminiferal ages on sediments on attained by the blueschists and eclogites. Our position top of the volcanic sequences (Tippit et al., 1981). on these two issues is as follows. 4 Amphibolites and greenschists, which constitute a 1 Phengite from high-pressure rocks is commonly metamorphic sole welded to the base of the Semail contaminated with excess Ar (e.g. Sherlock et al., ophiolite, formed between 96 and 89 Ma, as indicated 1999). The geological significance of 40Ar/39Ar ages for by 40Ar/39Ar and K–Ar data on hornblende, muscovite phengite should be considered suspect unless supported and biotite from these rocks (Lanphere, 1981; Hacker by age data obtained using other geochronological et al., 1997). These data also show that the amphibo- techniques. lites cooled through 550 uC between 95 and 93 Ma, 2 Pressures estimated for the As-Sifah high-P/T rocks indicating rapid cooling that is characteristic of normal using empirical geobarometers or thermodynamic oceanic crust (Hacker, 1994; Hacker et al., 1997). calculations of pressure-sensitive equilibria should be 5 Final emplacement of the Semail ophiolite onto the considered suspect unless they are consistent with the Oman continental margin took place between the end P–T stability ranges of the mineral assemblages in these of the Santonian and the early Maastrichtian (84 & rocks and all interlayered cofacial lithologies. 74.5 Ma), as indicated by the Turonian to Santonian The data most relevant to constraining the tectonic age of the Muti Formation (Le Me´tour et al., 1986) evolution of NE Oman are listed below. which structurally underlies the ophiolite, and by the Maastrichtian age of the laterites and conglomerates unconformably overlying it (Coleman, 1981; Hopson Stratigraphic and geochronological constraints et al., 1981). 1 The garnet blueschists and epidote amphibolites of 6 High-P/T metamorphism of the upper plate zone C appear to have the same protoliths as the (Regions I & II) of Saih Hatat took place in the Late crossite–epidote schists and greenschists of zone A, as Cretaceous, probably between 80 and 70 Ma. This is indicated by the geochemical data of El-Shazly et al. supported by 40Ar/39Ar ages for metasedimentary (1994, 1997). The persistent occurrence of Cpx-rich rocks from the Ruwi me´lange (El-Shazly & Lanphere, granofels within the mafic unit, as well as layers of 1992) and white micas from calcschists in Region II quartz–mica schists, metapelites and calcareous mica (Miller et al., 1998a). Although the interpretation of schists in all three metamorphic zones, suggests that the 40Ar/39Ar ages is fraught with uncertainties associated same units occur throughout Region III, and were with applying this technique to high-P/T rocks (as folded, boudinaged and metamorphosed under differ- pointed out earlier), these ages are considered reason- ent P–T conditions in the different zones. Whether the able because: (i) they are consistent with the Turonian protoliths of these rocks are pre-Permian (e.g. Glennie to Santonian age assigned to the Muti Formation, the et al., 1974) or Permian (Le Me´tour et al., 1986), they youngest unit in the upper plate metamorphosed under represent an integral part of the continental basement high-P/T conditions; (ii) the ages overlap with the or shelf of Oman. The suggestions of Michard et al. timing of ophiolite emplacement, a likely cause of high- 242 A. K. EL-SHAZLY ET AL.

P/T metamorphism (e.g. Goffe´ et al., 1988; El-Shazly, 100u C isotherm between 57 and 40 Ma (Poupeau et al., 1994, 1995); and (iii) the ages are consistent with the 1998). stratigraphic ages of the oldest unmetamorphosed sedimentary rocks that overlie the ophiolitic sequence Petrological constraints unconformably (e.g. Coleman, 1981). The apparent success of 40Ar/39Ar dating of the high-P/T meta- 1 The amphibolites of the metamorphic sole formed at morphism of the upper plate rocks as opposed to its a temperature of 775–875 uC (Searle & Malpas, 1980; failure with the lower plate eclogites can be explained Ghent & Stout, 1981) and a minimum pressure of by the considerably lower pressure of metamorphism of 8.5 kbar (based on the occurrence of kyanite in a garnet the former (3–6.5 kbar for Region I and 6.5–9 kbar for amphibolite; Gnos, 1998). Pressure estimates of Region II; see below), and the apparent tendency of 11–13 kbar (Gnos, 1998) are fraught with uncertainties white mica to incorporate excess Ar at higher pressures as they are based on intersections of thermodynami- (e.g. Sherlock et al., 1999). Following this line of cally calculated stable and metastable equilibria in the reasoning, a K–Ar age of 95¡4 Ma on white mica system Na2O–CaO–MgO–Al2O3–SiO2–H2O (NCMASH). from a calcschist from the deeper units of Region II 2 The Ruwi me´lange of Region I, Saih Hatat, was (Montigny et al., 1988) is considered unlikely to be metamorphosed at 280–320 uC and 3–6.5 kbar. Peak geologically meaningful and is attributed to excess Ar. P–T conditions of metamorphism of the deepest units 7 Eclogite facies metamorphism in the lower plate of in the upper plate (Region II) are estimated at Saih Hatat (Region III) took place shortly before ¡ 380–430 uC and 6.5–9 kbar (El-Shazly, 1994, 1995, 78 2 Ma, the time at which phengite cooled through 1996). Higher pressure estimates, such as those of Vidal its blocking temperature for Sr (500 uC). This inter- 40 39 et al. (1992), are suspect (cf. El-Shazly, 1996). All of pretation is compatible with the Ar/ Ar maximum these units followed clockwise P–T paths (El-Shazly, ages obtained for phengite from all three metamorphic 1994, 1995). zones (El-Shazly & Lanphere, 1992; Miller et al., 1999), 3 The lower plate rocks display an eastward increase in and overlaps with the timing of high-P/T metamorph- metamorphic grade from 400–460 uC, 6.5–8.5 kbar for ism in the upper plate. zone A to 500–580 uC, 12–16 kbar for zone C. Higher 8 The eclogite facies rocks cooled through 210 uCatc. pressure estimates for the As-Sifah eclogites (e.g. Searle 68¡2 Ma, and through the 100 uC isotherm at c. et al., 1994) based on empirical geobarometers are 54¡3 Ma, as indicated by fission track ages on zircon inconsistent with phase relations in interlayered and apatite, respectively (Saddiqi et al., 1995; Poupeau cofacial metapelites and quartz–mica schists (El- et al., 1998). Apatite fission track ages on upper plate Shazly et al., 1997; El-Shazly, 2001). As long as mineral rocks (Region II) indicate that they cooled through the assemblages indicative of such high pressures have not been found in these rocks, we prefer to remain conservative in our pressure estimates, making use of reliable geobarometers and petrogenetic grid calcula- tions. 4 All lower plate high-P/T metamorphic rocks fol- lowed clockwise P–T paths, with the eclogite facies rocks characterized by a segment of near-isothermal decompression between pressures of c. 12 kbar and c. 5 kbar at c. 500 uC, followed by another segment of near-isobaric cooling, as indicated by microthermo- metric measurements on solitary fluid inclusions in Cpx and quartz (El-Shazly & Sisson, 1999). 5 A pressure gap between the lower and upper plates exists and is most apparent in the eastern parts of Saih Hatat, where rocks metamorphosed at c. 6.5–9 kbar are juxtaposed against blueschist and eclogite facies rocks formed at 12–16 kbar. In western Saih Hatat, this gap is almost non-existent, where pressure estimates for Fig. 7. Temperature–time curves for the lower plate eclogite metasediments and metabasites of the upper plate facies rocks (filled squares) and the upper plate (Region II) Hijam and Hatat Formations overlap with those of schists (filled circles; broken curve), constrained by Rb–Sr (this zone A in the lower plate. However, the Hatat and study), fission track (FT) data (Saddiqi et al., 1995; Poupeau 40 39 Hijam units are characterized by the occurrence of et al., 1998) and Ar/ Ar data (El-Shazly & Lanphere, 1992; magnesioriebeckite/crossite and magnesiocarpholite, Miller et al., 1998a). Cooling rates for the lower plate rocks are given in italics. Cooling curves for both plates probably respectively, whereas zone A schists contain crossite converge at the 100 uC isotherm at c. 55 Ma, after which both and chloritoid but no Fe–Mg carpholite, which plates were exhumed at the same rate by erosion. suggests higher temperature (¡P) conditions for the FORMATION and EXHUMATION OF OMAN ECLOGITES 243 latter, and a possible pressure gap of < 1 kbar across by a foliation and bedding which are nearly parallel to the plate boundary in western Saih Hatat. their tectonic contact with the overlying Semail 6 High-P/T metamorphism affected the basement, ophiolite (e.g. Ghent & Stout, 1981). shelf and foreland basin units of Saih Hatat only. 8 The base of the Semail ophiolite is characterized by Rocks that structurally underlie the Semail ophiolite in mylonitized harzburgites with a foliation similar to that other tectonic windows do not show evidence of high- in the underlying amphibolites of the metamorphic sole P/T metamorphism. (Boudier & Coleman, 1981). The ophiolite itself is locally folded, with the Semail ‘thrust’ and Moho overturned in places (Gregory et al., 1998). Cooling rates 9 The Semail thrust has been reactivated as a low-angle Combining the fission track ages of Saddiqi et al. (1995) normal fault (e.g. Michard et al., 1994; Gregory et al., and Poupeau et al. (1998) with our Rb–Sr ages for the 1998), probably during the Tertiary events that led to eclogite facies rocks of As-Sifah leads us to suggest that the formation of the Saih Hatat, Jebel Akhdar and these rocks may have had two stages of cooling. During Hawasina domal structures. the first stage, which occurred between 78 and 68 Ma, the eclogite facies rocks cooled at a rate of c. Proposed tectonic model for the evolution of Saih Hatat 33 uCMax1. Between 68 and 54 Ma, the cooling rate decreased to 7–8 uCMax1 (Fig. 7). On the other In the Early Cretaceous at c. 96 Ma, crustal rocks of hand, the cooling rate of the upper plate units is the Semail ophiolite crystallized at the site of a fast, estimated at 7–11 uCMax1 based on the fission track mid-oceanic spreading centre located E and NE of the ages of Poupeau et al. (1998) and the assumption that Oman continental margin (Boudier & Nicolas, 1985; these rocks reached their peak temperature at c. 80 Ma. Fig. 8a). Several authors have suggested that this spreading centre was located in a back-arc basin (e.g. Lippard et al., 1986; Searle & Cox, 1999), but this is still Structural constraints debated (e.g. Coleman, 1981; Beurrier et al., 1989). 1 Lower plate units are deformed into a series of closed Between 96 and 95 Ma, a change in plate motion folds that verge to the north or north-east in zone A, caused Africa to begin its counterclockwise rotation and which become sheath folds verging ENE in zone C. towards Eurasia, which in turn resulted in intraoceanic These units are characterized by a penetrative foliation thrusting, marking the first stages of ophiolite obduc- that has been folded and transposed in the highest tion (Fig. 8b). Although this thrust was described by grade rocks of zone C. Searle et al. (1994) and Searle & Cox (1999) as a 2 A NE-trending mineral lineation which post-dates subduction zone, we prefer the term ‘intraoceanic the growth of most high-P/T porphyroblasts is well thrust’ to describe its low-angle nature (15u), in line developed in the lower plate rocks. Michard et al. with the terminology and conclusions of Cloos (1993). (1994) and Miller et al. (1999) associated this lineation The location of this detachment is controversial (cf. with sheath folding and top-to-the-NE shear sense Hacker & Gnos, 1997; Searle & Cox, 1999), but it must indicators, and interpreted it to have developed during have occurred close to (or at?) the site of the mid- extension. oceanic ridge (Boudier et al., 1988) in order to account 3 S–C fabrics and extensional crenulation cleavages for the high temperatures reached by the underplated that developed in the lower plate rocks under greens- crust as it was metamorphosed to amphibolites of the chist facies conditions indicate a down-to-the-NE sense sole. This metamorphism occurred at a depth of c. of shear (Michard et al., 1994; Searle et al., 1994; Miller 25 km to account for the P–T conditions estimated by et al., 1998a). Ghent & Stout (1981) and Gnos (1998). Fluids driven 4 In the northern parts of the upper plate (Region II), off from the underplated slab during upper amphibolite folds with S- or SW-directed vergence predominate. facies metamorphism infiltrated the mantle rocks of the The penetrative foliation of these units is axial planar overthrust plate, which may have triggered partial and nearly parallel to structural breaks within this plate melting in the hanging wall. Melts produced by this (e.g. Michard et al., 1994; Miller et al., 1999). process eventually penetrated the mantle and crustal 5 A NE-trending mineral lineation in Region II (upper sequence of the overthrust plate and the metamorphic plate) is related to SW-directed compressional fabrics sole, giving rise to the intrusive and extrusive rocks of (Michard et al., 1994). Miller et al. (1999) reported rare the middle Cenomanian to late Turonian ‘second shear bands at the structurally deepest levels of this magmatic event (M2)’ of the Semail ophiolite (e.g. Le plate which indicate top-to-the-NE transport. Me´tour et al., 1995) and leucogranitic dykes (Cox et al., 6 The low-angle ductile shear boundary between the 1999). Rocks formed during M2 are characterized by a lower and upper plates cross-cuts nappe structures in weak geochemical arc signature (Beurrier et al., 1989). both plates indicating relatively late movement along As the Semail ophiolite advanced towards the this boundary (Gregory et al., 1998). continental margin of Oman, the amphibolites, now 7 The metamorphic sole rocks which structurally welded to its base, were dragged with the ophiolite overlie the Saih Hatat high-P/T rocks are characterized across the Hawasina basin. These amphibolites cooled 244 A. K. EL-SHAZLY ET AL.

rapidly through 550 uC between 95 and 93 Ma (Hacker et al., 1997) as they were thrust onto colder rocks, which were in turn metamorphosed at lower tempera- ture conditions and welded to the base of the overriding slab to complete the formation of the metamorphic sole. As the ophiolite and metamorphic sole cooled and continued their advance towards the margin, con- tinental slope sediments of the Hawasina basin were scraped off the overridden plate and bulldozed in front of the leading edge of the advancing ophiolite, hence largely escaping metamorphism (Fig. 8b). As the ophiolite approached the continental margin of Oman, obduction continued because the continental crust was sufficiently thin along this margin (15 km for a 100 km thick lithosphere; Cloos, 1993). The impinge- ment of the ophiolite onto the continental margin of Oman c. 91–84 Ma resulted in the formation of a foreland bulge and basin. The foreland basin migrated inboard with time as it filled with sediments of the Turonian–Santonian Muti Formation, locally derived from the continental margin by mass wasting of the foreland bulge (Robertson, 1987a) and from the Hawasina basin by the bulldozing effect of the advancing ophiolite. This produced a tectonic me´lange in some parts of the Muti basin (Fig. 8c). Between 87 and 80 Ma, when the continental margin of Saih Hatat was being tectonically loaded by the Semail ophiolite, intracontinental thrusting took place, possibly triggered by a sudden increase in the rate of plate convergence. This event led to the development of a NE-dipping subduction zone and two discrete plates structurally beneath the Semail ophiolite: an upper plate consisting of folded and imbricated basement, continental shelf and foreland basin units (including the newly formed Muti tectonic me´lange), and a more coherent lower plate (Region III), both anchored to their lithospheric mantle (Fig. 8d). Subduction of the lower plate to depths > 15 km resulted in the Fig. 8. Simplified cartoon showing the stages of evolution of formation of blueschists and eclogites within its the continental margin of Saih Hatat with special emphasis on crust, which eventually became separated from the the formation and denudation of the high-pressure upper plate by a lithospheric mantle wedge thickening metamorphic rocks. (a) c. 96 Ma (Cenomanian). to the NE (Fig. 8d). In the meantime, the emplacement Crystallization of the Semail oceanic crust. (b) 96–93 Ma (Cenomanian). Intraoceanic detachment, formation of the of the ophiolite onto the continental margin caused metamorphic sole and M2 magmatism. (c) 91–84 Ma high-P/T metamorphism of the upper plate units, and (Turonian–Santonian). Deposition of the Muti Formation in a their simultaneous folding (with S- to SW-directed foreland basin. (d) 87–80 Ma (Santonian–Campanian). vergence) and thrusting (Fig. 8e). Intracontinental subduction and thickening of the Oman Between 80 and 78 Ma, when the leading edge of the continental margin. (e) c. 80 Ma (Campanian). Emplacement of the Semail ophiolite, metamorphism of the upper plate and lower plate reached a depth of 40–50 km, the continued subduction of the lower plate. (f) c.78Ma continental margin, with its thin crust (15 km), was (Campanian). Final emplacement of the ophiolite, choking of either consumed by subduction (lower plate) or over- the subduction zone and incipient tectonic exhumation of ridden by the ophiolite (upper plate). As convergence lower plate aided by ductile thinning of mantle wedge. (g) 78– 68 Ma (Campanian–Maastrichtian). Tectonic exhumation of continued, and thicker continental crust (c. 30 km) high-P/T rocks and juxtaposition of upper plate rocks against attempted to enter the subduction zone, it instead lower plate ones. Exhumation of the high-P/T rocks during the choked it due to its positive buoyancy (Cloos, 1993). Tertiary beyond stage (g) took place at a slower rate, primarily This in turn initiated the exhumation of the lower plate by erosion. Filled square, crossite–epidote schists of zone A; rocks, most likely along the trace of the subduction filled circle, zone B blueschists; filled diamond, zone C eclogites. Sm, Semail ophiolite; Hw, Hawasina sediments; ms, thrust as it was reactivated as a normal fault at c. metamorphic sole; mu, Muti Formation; RII, Region II; RIII, 78 Ma. As convergence and emplacement of the Region III. ophiolite continued simultaneously with the tectonic FORMATION and EXHUMATION OF OMAN ECLOGITES 245 exhumation of the lower plate rocks, the lithospheric apatite (Poupeau et al., 1998). Erosion, normal faulting mantle wedge separating the two plates (Regions II and and mass wasting resumed in Saih Hatat until the III) underwent ductile thinning (Fig. 8f). Tectonic eclogite facies rocks were finally exposed on the surface. exhumation and ductile thinning of this wedge both Our proposed tectonic model is based on ideas led to the development of the NE-vergent nappes presented in other models for the formation and (which telescoped the three metamorphic zones of the exhumation of high-P/T rocks (e.g. Philippot, 1990), lower plate), the NE-trending lineation and other and is very similar to that of Michard et al. (1994). extensional structures in the lower plate (e.g. Michard Although this model accounts for the geochronological et al., 1994). Eventually, these two processes led to the and petrological data available, and explains the juxtaposition of the upper plate rocks onto the lower pressure gaps between the upper and lower plates plate ones (Fig. 8g). and the differences in directions of fold vergence in By the end of the Campanian (c. 75 Ma), compres- both plates, it is still an oversimplified two-dimensional sion at the Oman margin waned, possibly as a result of speculative model. Models involving oblique subduc- the initiation of subduction of Tethyan oceanic litho- tion (Ave´ Lallement & Guth, 1990) and subsequent sphere along a N- to NE-dipping zone beneath the Iran extrusion should be entertained, particularly in the continental margin. With this new plate configuration, light of recent work suggesting late Maastrichtian to buoyancy forces continued to drive the lower plate early Tertiary compression along the SE margin of rocks towards the surface, initiating the updoming of Oman culminating in the emplacement of the Masirah the Saih Hatat area. Normal faulting within the upper island ophiolite (e.g. Immenhauser et al., 2000). More plate and the overlying Semail ophiolite (which broke research is therefore needed to: (i) identify the causes of into discrete blocks flanking Saih Hatat, Fig. 8g) was intracontinental subduction; (ii) provide structural the main mechanism for the exhumation of high-P/T evidence for the proposed mechanisms of exhumation rocks at this stage. In the Maastrichtian, the Semail of the high-P/T rocks; (iii) quantify the forces necessary ophiolite was exposed to subaerial weathering and to initiate exhumation; (iv) explain the development of erosion for a short period of time, marked by the a static greenschist facies overprint on the crossite– formation of laterites that overlie it unconformably epidote schists of zone A (El-Shazly & Lanphere, 1992), (Coleman, 1981). By the end of the Maastrichtian (c. when such an overprint is associated with extensional 68 Ma), the ophiolite had slid off the Saih Hatat proto- fabrics in zones B and C; and (v) account for the dome to the north and south, whereas the lower plate restriction of Fe–Mg carpholite grade high-P/T meta- eclogites had been exhumed to a depth of c.14kmas morphism to the Saih Hatat area, when it is attributed indicated by the fission track data of Saddiqi et al. to the tectonic loading of the continental margin by the (1995) on zircon (assuming that the geothermal ophiolite. gradients had relaxed from c. 10–12 uCkmx1, attained during high-P/T metamorphism, to c.15uCkmx1). During the late Palaeocene to early Eocene, a rise in ACKNOWLEDGEMENTS sea level caused the deposition of conglomerates, marls This study was supported by funds from the Deutsche and limestones on top of the Semail ophiolite in basins Forschungsgemeinschaft to Bro¨cker, and grants ES/97/ to the north, east and south-west of Saih Hatat, which 03 from the College of Science, Sultan Qaboos remained a tectonic high undergoing erosion (Le University to El-Shazly and NSF EAR-9725667 to Me´tour et al., 1995). At the same time, the high-P/T Hacker. Thorough reviews by Professor J. Selverstone metamorphic rocks continued their journey to the and Drs V. B. Sisson and M. Sullivan have improved surface, although at a much lower rate (c. the manuscript substantially. El-Shazly and Hacker are 0.4–0.8 mm yrx1; assuming that the geothermal gra- both grateful to the Ministry of Commerce and dient continued to relax to c.20uCkmx1). This Industry, particularly Mr. M. Kassim and Dr H. Al- process was driven primarily by erosion and mass Azry, for facilitating field work in Oman. Dr J. Rogers wasting (processes compatible with these rates, e.g. is thanked for drafting fig. 1, and T. Crane for help with Burbank & Beck, 1991; Burbank et al., 1996; Blythe, fig. 8. 1998), but may have also been aided by normal faulting in the upper plate. In the Eocene, subduction beneath the Makran continental margin of Iran had become REFERENCES well developed, causing the fully fledged development Arnaud, N. O. & Kelley, S. P., 1995. Evidence for excess argon of the domal structures of Saih Hatat, Jebel Akhdar during high pressure metamorphism in the Dora Maira Massif (western Alps, Italy), using an ultraviolet laser ablation and Hawasina windows. By 55 Ma, the eclogites had microprobe 40Ar/39Ar technique. Contributions to Mineralogy been exhumed to the same level as some units in Region and Petrology, 121, 1–11. II (Poupeau et al., 1998). Between 53 and 19 Ma, the Ave´ Lallement, H. & Guth, L. R., 1990. Role of extensional high-P/T rocks of both plates continued to be exhumed tectonics in exhumation of eclogites and blueschists in an oblique subduction setting: Northeastern Venezuela. Geology, by erosion. A Miocene–Pliocene compressional event 18, 950–953. culminating at c. 7–4 Ma caused minor heating of these Beurrier, M., Ohnenstetter, M., Cabanis, B., Lescuyer, J.L., rocks as indicated by fission track length data on Tegyey, M. & Le Me´tour, J., 1989. Geochemie des filons 246 A. K. EL-SHAZLY ET AL.

doleritiques et des roches volcaniques ophiolitiques de la nappe El-Shazly, A. K. & Coleman, R. G., 1990. Metamorphism in the de Semail: contraintes sur leur origine geotectonique au Oman Mountains in relation to the ophiolite emplacement. In: Cretace superieur. Bulletin Ge´ologique de France, 2 (8), 205– The Geology and Tectonics of the Oman Region. Geological 219. Society, London, Special Publication, 49 (eds Robertson, A. H. Blanckenburg, F. V., Villa, I. M., Baur, H., Morteani, G. & F., Searle, M. P. & Ries, A.), pp. 475–495. Geological Society, Steiger, R. H., 1989. Time calibration of a PT-path from the London. Western Tauern Window, Eastern Alps: the problem of closure El-Shazly, A. K. & Lanphere, M. A., 1992. Two high pressure temperatures. Contributions to Mineralogy and Petrology, 101, metamorphic events in NE Oman: evidence from 40Ar/39Ar 1–11. dating and petrological data. Journal of Geology, 100, 731–751. Blythe, A. E., 1998. Active tectonics and ultrahigh-pressure El-Shazly, A. K. & Liou, J. G., 1991. Glaucophane chloritoid rocks. In: When Continents Collide: Geodynamics and bearing assemblages from NE Oman: petrologic significance Geochemistry of Ultrahigh-Pressure Rocks (eds Hacker, B. R. and a petrogenetic grid for high P/T metapelites. Contributions & Liou, J. G), pp. 141–160. Kluwer Academic Publishers, to Mineralogy and Petrology, 107, 180–201. Dordrecht. El-Shazly, A. K. & Sisson, V. B., 1999. Retrograde evolution of Boudier, F. & Coleman, R. G., 1981. Cross section through the eclogites and blueschists from NE Oman: evidence from fluid peridotite in the Semail ophiolite, southeastern Oman. Journal inclusions and petrological data. Chemical Geology, 154, 193– of Geophysical Research, 86, 2573–2592. 223. Boudier, F. & Nicolas, A., 1985. Harzburgites and lherzolites in El-Shazly, A. K., Coleman, R. G. & Liou, J. G., 1990. Eclogites subtypes in ophiolitic and oceanic environments. Earth and and blueschists from NE Oman: petrology and P–T evolution. Planetary Science Letters, 76, 84–92. Journal of Petrology, 31, 629–666. Boudier, F., Ceulleneer, G. & Nicolas, A., 1988. Shear zones, El-Shazly, A. K., Worthing, M. A., Jayawardane, J. & Varne, R., thrusts and related magmatism in the Oman ophiolite: 1994. Geochemistry of metamorphosed mafic rocks from Saih initiation of thrusting on an oceanic ridge. Tectonophysics, Hatat: insights into the pre-obduction history of NE Oman. 51, 275–296. Journal of the Geological Society of London, 151, 999–1016. Bro¨cker, M. & Franz, L., 1998. Rb–Sr isotope studies on Tinos El-Shazly, A. K., Worthing, M. A. & Liou, J. G., 1997. Island (Cyclades, Greece): additional time constraints for Interlayered eclogites, blueschists and epidote amphibolites metamorphism, extent of infiltration controlled overprinting from NE Oman: a record of protolith compositional control and deformational activity. Geological Magazine, 135, 369– and limited fluid infiltration. Journal of Petrology, 38, 1461– 382. 1487. Burbank, D. & Beck, R., 1991. Rapid long-term rates of Gebauer, D., Schertl, H.-P., Brix, M. & Schreyer, W., 1997. denudation. Geology, 19, 1169–1172. 35 Ma old ultrahigh-pressure metamorphism and evidence for Burbank, D., Leland, J., Fielding, E., et al., 1996. Bedrock very rapid exhumation in the Dora Maira Massif, Western incision, rock uplift and threshold hillslopes in the north- Alps. Lithos, 41, 5–24. western Himalayas. Nature, 379, 505–510. Ghent, E. D. & Stout, M. Z., 1981. Metamorphism at the base of Chemenda, A. I., Mattauer, M. & Bokun, A. N., 1996. the Semail ophiolite, Oman. Journal of Geophysical Research, Continental subduction and a mechanism for exhumation of 86, 2557–2571. high pressure metamorphic rocks: new modelling and field Glennie, K. W., Boeuf, M. G. A., Hughes Clark, M. H. W., data from Oman. Earth and Planetary Science Letters, 143, Moody-Stuart, M., Pilaar, W. F. & Reinhardt, B. M., 173–182. 1974. Geology of the Oman mountains. Verhandelingen van Cloos, M., 1993. Lithospheric buoyancy and collisional orogen- het Koninklijk Nederlands Geologischs Mijnbouwkundig esis: subduction of oceanic plateaus, continental margins, Genootschap, 31, 1–423. island arcs, spreading ridges, and seamounts. Geological Goffe´, B., Michard, A., Kienast, J. R. & Le Mer, O., 1988. A case Society of America, Bulletin, 105, 715–737. of obduction related high P low T metamorphism in upper Coleman, R. G., 1981. Tectonic setting for ophiolite obduction in crustal nappes, Arabian continental margin, Oman: P–T paths Oman. Journal of Geophysical Research, 86, 2497–2508. and kinematic interpretation. Tectonophysics, 151, 363–386. Cox, J., Searle, M. P. & Pedersen, R., 1999. The petrogenesis of Gnos, E., 1998. Peak metamorphic conditions of garnet leucogranitic dykes intruding the northern Semail ophiolite, amphibolites beneath the Semail ophiolite: implications for United Arab Emirates: field relationships, geochemistry and an inverted pressure gradient. International Geology Review, Sr/Nd isotope systematics. Contributions to Mineralogy and 40, 281–304. Petrology, 137, 267–287. Gregory, R. T., Gray, D. R. & Miller, J. McL., 1998. Tectonics of Cumbest, R. J., Johnson, E. L. & Onstott, T. C., 1994. Argon the Arabian margin associated with the formation and composition of metamorphic fluids: implications for 40Ar/39Ar exhumation of high pressure rocks, Sultanate of Oman. geochronology. Geological Society of America, Bulletin, 106, Tectonics, 17, 657–670. 942–951. Hacker, B. R., 1994. Rapid emplacement of young oceanic Dallmeyer, R. D., Peucat, J. J., Hirajima, T. & Ohta, Y., 1990. lithosphere. Science, 265, 1563–1565. Tectonothermal chronology within a blueschist–eclogite com- Hacker, B. R. & Gnos, E., 1997. The conundrum of Semail: plex, west-central Spitsbergen, Svalbard: evidence from explaining the metamorphic history. Tectonophysics, 279, 215– 40Ar/39Ar and Rb–Sr mineral ages. Lithos, 24, 291–304. 226. Dalrymple, G. B. & Lanphere, M. A., 1974. 40Ar/39Ar age spectra Hacker, B. R. & Wang, Q. C., 1995. Ar/Ar geochronology of of some undisturbed terrestrial samples. Geochimica et ultrahigh-pressure metamorphism in central China. Tectonics, Cosmochimica Acta, 38, 359–372. 14, 994–1006. El-Shazly, A. K., 1994. Petrology of lawsonite, pumpellyite and Hacker, B. R., Mosenfelder, J. L. & Gnos, E., 1997. Rapid sodic amphibole-bearing rocks from Saih Hatat, NE Oman. ophiolite emplacement constrained by geochronology and Journal of Metamorphic Geology, 12, 23–48. thermal considerations. Tectonics, 15, 1230–1247. El-Shazly, A. K., 1995. Petrology of Fe–Mg carpholite-bearing Hammerschmidt, K. & Franz, G., 1992. Retrograde evolution metasediments from NE Oman. Journal of Metamorphic of eclogites: evidence from microstructures and 40Ar/39Ar Geology, 13, 379–396. white mica dates, Munchberg Massif, northern Bavaria. El-Shazly, A. K., 1996. Petrology of Fe–Mg carpholite-bearing Contributions to Mineralogy and Petrology, 111, 113–125. metasediments from NE Oman: reply to comments by Vidal Hannula, K. A. & McWilliams, M. O., 1995. Reconsideration of and Theye. Journal of Metamorphic Geology, 14, 386–397. the age of blueschist facies metamorphism on the Seward El-Shazly, A. K., 2000. Are pressures in blueschists and eclogites Peninsula, Alaska, based on phengite 40Ar/39Ar results. overestimated? The case from Oman. Lithos, in press. Journal of Metamorphic Geology, 13, 125–139. FORMATION and EXHUMATION OF OMAN ECLOGITES 247

Hopson, C. A., Coleman, R. G., Gregory, R. T., Pallister, J. S. & high-pressure metamorphic terrane, Oman. In: Exhumation Bailey, E. H., 1981. Geologic section through the Muscat–Ibra Processes: Normal Faulting, Ductile Flow and Erosion. transect, southeastern Oman Mountains. Journal of Geological Society, London, Special Publication, 154 (eds Geophysical Research, 86, 2527–2544. Ring, U., Brandon, M. T., Lister, G. S. & Willet, S. D.), pp. Hunziker, J. C., Desmons, J. & Martinotti, G., 1989. Alpine 241–260. Geological Society, London. thermal evolution in the central and western Alps. In: Alpine Montigny, R., Le Me´r, O., Thuizat, R. & Whitechurch, H., 1988. Tectonics. Geological Society, London, Special Publication,45 K–Ar and 40Ar/39Ar study of metamorphic rocks associated (eds Coward, M. P., Dietrich, D. & Park, R. G.), pp. 353–367. with the Oman ophiolite: tectonic implications. Tectono- Geological Society, London. physics, 151, 345–362. Immenhauser, A., Schreurs, G., Gnos, E., Oterdoom, H. W. & Oberha¨nsli, R., Wendt, A. S., Goffe´, B. & Michard, A., 1999. Hartmann, B., 2000. Late Paleozoic to Neogene geodynamic Detrital chromites in metasediments of the East Arabian evolution of the northeastern Oman margin. Geological continental margin in the Saih Hatat area: constraints for the Magazine, 137, 1–18. paleogeographic setting of the Hawasina and Semail basins. Inger, S., Ramsbotham, W., Cliff, R. A. & Rex, D. C., 1996. International Journal of Earth Sciences (Geologisches Metamorphic evolution of the Sesia-Lanzo Zone, Western Raundschau), 88, 13–25. Alps: time constraints from multi-system geochronology. Philippot, P., 1990. Opposite vergence of nappes and crustal Contributions to Mineralogy and Petrology, 126, 152–168. extension in the French–Italian western Alps. Tectonics, 9, Kirschner, D. L., Cosca, M. A., Masson, H. & Hunziker, J. C., 1143–1164. 1996. Staircase 40Ar/39Ar spectra of fine-grained white mica: Poupeau, G., Saddiqi, O., Michard, A., Goffe´, B. & Oberha¨nsli, timing and duration of deformation and empirical constraints R., 1998. Late thermal evolution of the Oman Mountains on argon diffusion. Geology, 24, 747–750. subophiolitic windows: apatite fission track thermochronol- Lanphere, M. A., 1981. K–Ar ages of metamorphic rocks at the ogy. Geology, 26, 1139–1142. base of the Semail ophiolite, Oman. Journal of Geophysical Purdy, J. W. & Ja¨ger, E., 1976. K–Ar ages on rock-forming Research, 86, 2777–2782. minerals from the central Alps. Memoria Instituto Geologica e Lanphere, M. A. & Dalrymple, G. B., 1976. Identification of Mineralogica, Universita di Padova, 30. excess 40Ar by the 40Ar/39Ar age spectrum technique. Earth and Reddy, S. M., Kelley, S. P. & Wheeler, J., 1996. A 40Ar/39Ar laser Planetary Science Letters, 32, 141–148. probe study of micas from the Sesia Zone, Italian Alps: Le Me´tour, J., de Gramont, X. & Villey, M., 1986. Geological implications for metamorphic and deformation histories. Map of Masqat and Quryat, Sheets NF40-4A, NF40-4D. Scale Journal of Metamorphic Geology, 14, 493–508. 1 : 100, 000. Ministry of Petroleum and Minerals, Directorate Robertson, A. H. F., 1987a. Upper Cretaceous Muti Formation: General of Minerals, Sultanate of Oman. transition of a Mesozoic carbonate platform to a foreland Le Me´tour, J., Michel, J. C., Be´chennec, F., Platel, J. P. & Roger, basin in the Oman Mountains. Sedimentology, 34, 1123–1142. J., 1995. Geology and Mineral Wealth of the Sultanate of Oman. Robertson, A. H. F., 1987b. The transition from a passive margin Ministry of Petroleum and Minerals Geological Documents, to an Upper Cretaceous foreland basin related to ophiolite Sultanate of Oman. emplacement in the Oman Mountains. Geological Society of Le Me´tour, J., Rabu, D., Tegyey, M., Bechennec, F., Beurrier, America, Bulletin, 99, 633–653. M. & Villey, M., 1990. Subduction and obduction: two stages Ruffet, G., Fe´raud, G., Balle´vre, M. & Kienast, J. R., 1995. in the Eo-Alpine tectonometamorphic evolution of the Oman Plateau ages and excess argon in phengites: an 40Ar/39Ar laser Mountains. In: The Geology and Tectonics of the Oman Region. probe study of Alpine micas (Sesia Zone, Western Alps, Geological Society, London, Special Publication, 49 (eds northern Italy). Chemical Geology, 121, 327–343. Robertson, A. H. F., Searle, M. P. & Ries, A. C.), pp. 327– Saddiqi, O., Poupeau, G., Michard, A., Goffe´, B. & Oberha¨nsli, 339. Geological Society, London. R., 1995. Exhumation des roches metamorphiques HP-LT Li, S., Wang, S., Chen, Y., et al., 1994. Excess Ar in phengite d’Oman: Datation par traces de fission sur zircons. Paris, from eclogite: evidence from dating of eclogite by Sm–Nd, Academie des Sciences, Comptes Rendus, 320, 1071–1077. Rb–Sr and 40Ar/39Ar methods. Chemical Geology Isotope Scaillet, S., Fe´raud, G., Balle`vre, M. & Amouric, M., 1992. Mg/ Geoscience, 112, 343–350. Fe and [(Mg,Fe)Si-Al2] control on Ar behaviour in high Lippard, S. J., 1983. Cretaceous high pressure metamorphism in pressure white micas: a 40Ar/39Ar continuous laser probe study NE Oman and its relationship to subduction and ophiolite from the Dora Maira nappe of the internal western Alps, Italy. nappe emplacement. Journal of the Geological Society, London, Geochimica et Cosmochimica Acta, 56, 2851–2872. 140, 97–104. Schermer, E. R., Lux, D. R. & Burchfiel, B. C., 1990. Lippard, S. J., Shelton, A. W. & Gass, I. G., 1986. The Ophiolite Temperature–time history of subducted continental crust, of Northern Oman. Blackwell Scientific Publications, Oxford. Mount Olympos Region, Greece. Tectonics, 9, 1165–1195. McDougall, I. & Harrison, T. M., 1988. Geochronology Searle, M. P. & Cooper, 1986. Structure of the Hawasina and Thermochronology by the 40Ar/39Ar Method. Oxford Window culmination, central Oman Mountains. Transactions Monographs, Geology and Geophysics, 9. Oxford University of the Royal Society of Edinburgh: Earth Sciences, 77, 143–156. Press, Oxford. Searle, M. P. & Cox, J., 1999. Tectonic setting, origin, and Michard, A., Bouche´z, J. L. & Ouazzani-Touhami, M., 1984. obduction of the Oman ophiolite. Geological Society of Obduction related planar and linear fabrics in Oman. Journal America, Bulletin, 111, 104–122. of Structural Geology, 6, 39–49. Searle, M. P. & Malpas, J., 1980. Structure and metamorphism of Michard, A., Goffe´, B. & Ouazzani-Touhami, M., 1983. rocks beneath the Semail ophiolite of Oman and their tectonic Obduction related high pressure, low temperature metamorph- significance in ophiolite obduction. Transactions of the Royal ism in upper crustal materials, Muscat, Oman. Terra Cognita, Society of Edinburgh: Earth Sciences, 71, 247–262. 3, IA22–23. Searle, M. P., Lippard, S. J., Smewing, J. D. & Rex, D. C., 1980. Michard, A., Goffe´, B., Saddiqi, O., Oberha¨nsli, R. & Wendt, A. Volcanic rocks beneath the Semail ophiolite nappe in the S., 1994. Late Cretaceous exhumation of the Oman blueschists northern Oman mountains and their significance in the and eclogites: a two-stage extensional mechanism. Terra Nova, Mesozoic evolution of Tethys. Journal of the Geological 6, 404–413. Society, London, 137, 589–604. Miller, J. M. L., Gray, D. R. & Gregory, R. T., 1999. Searle, M. P., Waters, D. J., Martin, H. N. & Rex, D. C., 1994. Exhumation of high pressure rocks in northeastern Oman. Structure and metamorphism of blueschist–eclogite facies Geology, 26, 235–238. rocks from the northeastern Oman Mountains. Journal of Miller, J. M. L., Gregory, R. T. & Gray, D. R., 1999. Geological the Geological Society, London, 151, 555–576. and geochronological constraints on the exhumation of a Sherlock, S., Kelley, S., Inger, S., Harris, N. & Okay, A., 1999. 248 A. K. EL-SHAZLY ET AL.

40Ar/39Ar and Rb–Sr geochronology of high-pressure meta- biostratigraphy of sediments in the volcanic unit of the Semail morphism and exhumation history of the Tavsanli Zone, NW ophiolite. Journal of Geophysical Research, 86, 2756–2762. Turkey. Contributions to Mineralogy and Petrology, 137, 46– Tonarini, S., Villa, I. M., Oberli, F., et al., 1993. Eocene age of 58. eclogite metamorphism in Pakistan Himalaya: implications for Sisson, V. B. & Onstott, T. C., 1986. Dating blueschist India–Eurasia collision. Terra Nova, 5, 13–20. metamorphism: a combined 40Ar/39Ar and electron microp- Vidal, O., Goffe´, B. & Theye, T., 1992. Experimental study of the robe approach. Geochimica et Cosmochimica Acta, 50, 2111– stability of sudoite and magnesiocarpholite and calculation of 2117. a new petrogenetic grid for the system FeO–MgO–Al2O3– Steiger, R. H. & Ja¨ger, E., 1977. Subcommission on geochronol- SiO2–H2O. Journal of Metamorphic Geology, 10, 603–614. ogy: convention on the use of decay constants in geo- and Wendt, A. S., D’Arco, P., Goffe´, B. & Oberha¨nsli, R., 1993. cosmochronology. Earth and Planetary Science Letters, 36, Radial cracks around a-quartz inclusions in almandine: 359–362. constraints on the metamorphic history of the Oman Tilton, G. R., Hopson, C. A. & Wright, J. E., 1981. Uranium– Mountains. Earth and Planetary Science Letters, 114, 449–461. 40 39 lead isotopic ages of the Semail ophiolite, Oman, with Wijbrans, J. R. & McDougall, I., 1986. Ar/ Ar dating of white applications to Tethyan Ocean Ridge Tectonics. Journal of micas from an Alpine high pressure metamorphic belt on Geophysical Research, 86, 2756–2762. Naxos (Greece): the resetting of the argon isotopic system. Tilton, G. R., Schreyer, W. & Schertl, H.-P., 1991. Pb–Sr–Nd Contributions to Mineralogy and Petrology, 93, 187–194. isotopic behaviour of deeply subducted crustal rocks from the York, D., 1969. Least squares fitting of a straight line with Dora Maira massif, western Alps, Italy II: what is the age of correlated errors. Earth and Planetary Science Letters, 5, 320– ultrahigh pressure metamorphism? Contributions to 324. Mineralogy and Petrology, 108, 22–33. Tippit, P. R., Pessagno, E. A. & Smewing, J. D., 1981. The Received 11 February 2000; revision accepted 20 September 2000